Medical imaging, a cornerstone of disease diagnosis and treatment planning, faces the hurdles of subjective interpretation and reliance on specialized expertise. Deep learning algorithms show improvements in automatin...
详细信息
Emotions have a significant impact on how people make decisions. Due to its potential applications in various fields, emotion intensity detection has attracted a lot of attention recently. Several methods have been pr...
详细信息
Deep neural networks (DNNs) are crucial in autonomous driving systems (ADSs) for tasks like steering control, but model inaccuracies, biased training data, and incorrect runtime parameters can compromise their reliabi...
详细信息
Deep neural networks (DNNs) are crucial in autonomous driving systems (ADSs) for tasks like steering control, but model inaccuracies, biased training data, and incorrect runtime parameters can compromise their reliability. Metamorphic testing (MT) enhances reliability by generating follow-up tests from mutated DNN source inputs, identifying inconsistencies as defects. Various MT techniques for ADSs include generative/transfer models, neuron-based coverage maximization, and adaptive test selection. Despite these efforts, significant challenges remain, including the ambiguity of neuron coverage’s correlation with misbehaviour detection, a lack of focus on DNN critical pathways, inadequate use of search-based methods, and the absence of an integrated method that effectively selects sources and generates follow-ups. This paper addresses such challenges by introducing DeepDomain, a grey-box multi-objective test generation approach for DNN models. It involves adaptively selecting diverse source inputs and generating domain-oriented follow-up tests. Such follow-ups explore critical pathways, extracted by neuron contribution, with broader coverage compared to their source tests (inter-behavioural domain) and attaining high neural boundary coverage of the misbehaviour regions detected in previous follow-ups (intra-behavioural domain). An empirical evaluation of the proposed approach on three DNN models used in the Udacity self-driving car challenge, and 18 different MRs demonstrates that relying on behavioural domain adequacy is a more reliable indicator than coverage criteria for effectively guiding the testing of DNNs. Additionally, DeepDomain significantly outperforms selected baselines in misbehaviour detection by up to 94 times, fault-revealing capability by up to 79%, output diversity by 71%, corner-case detection by up to 187 times, identification of robustness subdomains of MRs by up to 33 percentage points, and naturalness by two times. The results confirm that stat
Recommender systems play an essential role in decision-making in the information age by reducing information overload via retrieving the most relevant information in various applications. They also present great oppor...
详细信息
The transition from traditional energy or electrical grids to smart energy or electrical grids has significantly transformed energy management. This evolution emphasizes decentralization, efficiency, and sustainabilit...
详细信息
The transition from traditional energy or electrical grids to smart energy or electrical grids has significantly transformed energy management. This evolution emphasizes decentralization, efficiency, and sustainability in energy systems. However, it also introduces numerous risks, including cyber-physical system vulnerabilities and challenges in energy trading. The application of blockchain and Machine Learning (ML) offers potential solutions to these issues. Blockchain enhances energy transactions by making them safer, more transparent, and tamper-proof, while ML optimizes grid performance by improving predictions, fault detection, and anomaly identification. This systematic review examines the application of blockchain and ML in peer-to-peer (P2P) energy trading within smart grids and analyzes how these technologies complement each other in mitigating risks and enhancing the efficiency of smart grids. Blockchain enhances security by providing privacy for transactions and maintaining immutable records, while ML predicts market trends, identifies fraudulent activities, and ensures efficient energy use. The paper identifies critical challenges in smart grids, such as unsecured communication channels and vulnerabilities to cyber threats, and discusses how blockchain and ML address these issues. Furthermore, the study explores emerging trends, such as lightweight blockchain systems and edge computing, to overcome implementation challenges. A new architecture is proposed, integrating blockchain with ML algorithms to create resilient, secure, and efficient energy trading markets. The paper underscores the need for global standardization, improved cybersecurity measures, and further research into how blockchain and ML can revolutionize smart grids. This study integrates current knowledge with a forward-looking perspective, providing valuable insights for researchers, policymakers, and stakeholders in the energy sector to collaboratively build a future of efficient and int
Diabetic Retinopathy is a disease,which happens due to abnormal growth of blood vessels that causes spots on the vision and vision *** techniques are applied to identify the disease in the early stage with different m...
详细信息
Diabetic Retinopathy is a disease,which happens due to abnormal growth of blood vessels that causes spots on the vision and vision *** techniques are applied to identify the disease in the early stage with different methods and *** Learning(ML)techniques are used for analyz-ing the images andfinding out the location of the *** restriction of the ML is a dataset size,which is used for model *** problem has been overcome by using an augmentation method by generating larger datasets with multidimensional *** models are using only one augmentation tech-nique,which produces limited features of dataset and also lacks in the association of those data during DR detection,so multilevel augmentation is proposed for *** proposed method performs in two phases namely integrated aug-mentation model and dataset correlation(***).It eliminates overfit-ting problem by considering relevant *** method is used for solving the Diabetic Retinopathy problem with a thin vessel identification using the UNET *** based image segmentation achieves 98.3%accuracy when com-pared to RV-GAN and different UNET models with high detection rate.
The nonlinear filtering problem has enduringly been an active research topic in both academia and industry due to its ever-growing theoretical importance and practical *** main objective of nonlinear filtering is to i...
详细信息
The nonlinear filtering problem has enduringly been an active research topic in both academia and industry due to its ever-growing theoretical importance and practical *** main objective of nonlinear filtering is to infer the states of a nonlinear dynamical system of interest based on the available noisy measurements. In recent years, the advance of network communication technology has not only popularized the networked systems with apparent advantages in terms of installation,cost and maintenance, but also brought about a series of challenges to the design of nonlinear filtering algorithms, among which the communication constraint has been recognized as a dominating concern. In this context, a great number of investigations have been launched towards the networked nonlinear filtering problem with communication constraints, and many samplebased nonlinear filters have been developed to deal with the highly nonlinear and/or non-Gaussian scenarios. The aim of this paper is to provide a timely survey about the recent advances on the sample-based networked nonlinear filtering problem from the perspective of communication constraints. More specifically, we first review three important families of sample-based filtering methods known as the unscented Kalman filter, particle filter,and maximum correntropy filter. Then, the latest developments are surveyed with stress on the topics regarding incomplete/imperfect information, limited resources and cyber ***, several challenges and open problems are highlighted to shed some lights on the possible trends of future research in this realm.
Semantic segmentation of driving scene images is crucial for autonomous *** deep learning technology has significantly improved daytime image semantic segmentation,nighttime images pose challenges due to factors like ...
详细信息
Semantic segmentation of driving scene images is crucial for autonomous *** deep learning technology has significantly improved daytime image semantic segmentation,nighttime images pose challenges due to factors like poor lighting and overexposure,making it difficult to recognize small *** address this,we propose an Image Adaptive Enhancement(IAEN)module comprising a parameter predictor(Edip),multiple image processing filters(Mdif),and a Detail Processing Module(DPM).Edip combines image processing filters to predict parameters like exposure and hue,optimizing image *** adopt a novel image encoder to enhance parameter prediction accuracy by enabling Edip to handle features at different *** strengthens overlooked image details,extending the IAEN module’s *** the segmentation network,we integrate a Depth Guided Filter(DGF)to refine segmentation *** entire network is trained end-to-end,with segmentation results guiding parameter prediction optimization,promoting self-learning and network *** lightweight and efficient network architecture is particularly suitable for addressing challenges in nighttime image *** experiments validate significant performance improvements of our approach on the ACDC-night and Nightcity datasets.
Spiking Neural Networks (SNNs) can be a suitable alternative to deep artificial neural networks due to efficient computation and low-power event-driven information analysis. Nevertheless, SNN cannot compete with Artif...
详细信息
Machine learning (ML) with data analysis has many successful applications and is widely employed daily. Additionally, they have played a significant role in combating the global coronavirus (COVID-19) outbreak. Intern...
详细信息
暂无评论